A chess server and an insurance prospecting tool rank in the nonprofit top 13

A chess server and an insurance prospecting tool rank in the nonprofit top 13

Nonprofit & Social Impact holds 229 providers and 1,221 APIs, with a median Kin Score of 18.0 and a median Agent Readiness of 3.1. That is a low bar, and the top of the list flatters it. Five of the top 13 carry no fundraising, donation, giving, grant, charity, philanthropy, donor or impact tag. Lichess ranks 12th, and its record describes an online chess server. LeO ranks 13th and sells prospecting software to commercial insurance professionals.

The vertical by function

Function Provider Kin Score Agent Readiness APIs
Fundraising GoFundMe 68.8 43.0 72
Fundraising Givebutter 65.9 55.3 21
Donor and organizing CRM NationBuilder 60.1 32.9 41
Donor and organizing CRM Blackbaud 49.6 30.6 313
Giving infrastructure Goodstack 59.7 41.3 42
Giving infrastructure Chariot 55.2 52.1 29
Sector data Plinth US Grants Data 58.9 49.0 5
Sector data Candid 35.7 47.7 9

Three patterns

The tag means two things. 201 of the 229 carry a Non-Profit or Nonprofit tag, and that covers organizations that are nonprofits and vendors that sell to them. publiq leads the vertical at 69.9 as a Belgian non-profit publishing culture and leisure data. Lichess, MusicBrainz and IAB Tech Lab qualify on legal status too. Levitate and LeO qualify because their descriptions name nonprofits among the customers they serve. The tag is accurate in most of these cases, and the vertical it feeds is still the wrong home for them. The membership rule is ours.

Separating the sector improves it only a little. The 119 members with one of those sector tags have a median of 20.8 and 6.0. The other 110 sit at 12.9 and 3.1. Even that cut is rough, since StreamElements, a platform for live streamers, gets in on a Donations tag. The core of the sector is stronger than the vertical’s headline number, and it is still low. 83 members score zero on agents.

Agent scores run ahead of contracts here. Zeffy is the extreme case, with a Kin Score of 13.2 and an Agent Readiness of 40.9. Its record shows an MCP server and scores 0.0 on both contract quality and contract governance. Fundraise Up (25.1 and 40.9), Candid (35.7 and 47.7) and Donorbox (28.8 and 34.6) follow the same pattern, and each records an MCP server. A small fundraising tool can add an agent front door faster than it can publish and maintain a contract. That is a reasonable order of work, provided the contract does eventually arrive.

Where to start

For accepting donations, read Givebutter first, since it has the better agent score of the two leaders. For embedding giving in a product, compare Goodstack with Chariot. For US grant data, Plinth US Grants Data and Candid cover overlapping ground from different directions. Blackbaud has the largest surface in the sector at 313 APIs.

Takeaway

This vertical needs a stricter membership rule from us more than it needs a better average. Charities should read the fundraising rows and ignore the chess server, however good its API is.

See the vertical at apis.io/industries/nonprofit-social-impact/.

← Statsig's Experiments API lists seven URLs twice, and the two contracts disagree
Junction's Order API offers idempotency on 3 of 36 operations, and one of those is in closed beta →